### Table of Contents

# HF dummy load power meter homebrew

## Introduction

For measuring the output power of a transceiver, a measurement range from one watt to a few hundred watts and an accuracy of a few percent is more than enough.

This measurement range and accuracy can easily be achieved by connecting the transceiver to a dummy load and measuring the voltage across the dummy load with a single-phase rectifier. The impedance of the dummy load should match the output impedance of the transceiver. In practice, this is almost always 50 ohms.

The dummy load power meter principle is a good alternative to the inaccurate SWR meter with power scale and the accurate but expensive commercial power meter.

- Advantages
- Simple measurement principle
- Cheap and easy to build yourself
- Practically usable from less than a watt to more than a kilowatt
- High accuracy (a few percent)
- Digital Multimeter does not require calibration

- Disadvantages
- No live transmission measurement possible, transceiver must be disconnected from antenna
- Maximum measuring time depends on the heatsink used

On HF dummy load power meter theory you will find the theoretical description and the derivation of the formulas used for this measurement method.

## Basic diagram

The dummy load should have an SWR close to 1:1 over the entire frequency range used. Because of this low SWR, the DC output voltage gives an accurate indication of the power absorbed in the dummy load.

### Measuring a Digital Multi Meter

The simplest method for measuring the DC output voltage is an accurate digital multimeter (DMM). Using the formules below, it is easy to calculate the power absorbed in the Dummy load from this.

You don't need to calibrate anything when using a DMM. The accuracy of this method is directly within a few per cent.

### Standalone instrument

You can also build a standalone instrument that shows the measured power directly on a scale or display. This can be either analogue or digital.

#### Analogue

A good example of an analogue standalone power meter is the project by Z33T Mile gives a detailed build description with photos and also describes a simple and accurate method to calibrate the scale of your analogue meter in Watts.

- Benefits
- Power directly readable
- Simple construction
- high accuracy

- Disadvantages
- Requires customised scale
- A calibration procedure is required when making scales

#### Digital

If you want to build digital power readouts, there are many options, such as an Arduino. The Arduino can convert the measured DC voltage directly to power.

When measuring higher powers, the DC voltage rises rapidly (50 watts gives about 70 volts). Keep in mind that for higher powers, you need a voltage divider to avoid damaging the analogue input of the Arduino.

- Advantages
- Power can be read off directly
- accurate measurement
- No calibration required

- Disadvantages
- More complex and expensive construction
- programming knowledge required

## Formulas

To calculate the measured power, I use the formulas below. On HF dummy load power meter theory you can find the derivation of these formulas.

- `P` is dissipated power in Watts
- `Uo` is the measured DC output voltage in volts
- `Uk` is the knee voltage of the diode(s) used in volts
- `Ur` is maximum reverse voltage diode(s)

For a 50 Ohm dummy load:

`P = ((Uo + Uk)/10)^2`

`Uo = 10sqrtP - Uk`

`Ur = 20*sqrt(P) + Uk`

`Pmax = ((Ur - Uk)/20)^2`

For high values of Uo and Ur, the influence of Uk is negligible and can be omitted, simplifying the formulas.

`P = ((Uo)/10)^2`

`Uo = 10sqrtP`

`Ur = 20*sqrt(P)`

`Pmax = ((Ur)/20)^2`

## Diagram and components

The theoretical scheme with ideal compositions is straightforward. In practice, we have to take into account the practical properties of the components used.

### Dummy load

For this design, I used a 50 ohm, 250-watt dummy load with a frequency range of 0 - 3 GHz. These dummy load resistors can be bought for little on several major online platforms and are easy to mount on a heatsink.

**Please note**

- The HF connection lug is quite fragile, bend it up and down as little as possible.
- These resistors contain beryllium oxide. Do not saw or file these resistors as the dust released is toxic.

### Heatsink

The temperature of the dummy load used is allowed to reach 100 ºC before the maximum dissipated power drops. (See the graph in the datasheet).

When choosing a heatsink have looked in your junk box for suitable heatsink. This heatsink can dissipate 100 watts for about 3 minutes before the dummy load reaches 100 ºC. This is long enough for my measurements. For longer measurement time, you can opt for a larger heatsink and/or a cooling fan.

### Capacitors

Each type of capacitor has a maximum usable frequency. Above this frequency, the self-inductance of the capacitor plays an increasing role. The capacitor behaves like an inductor. For high-capacity capacitors, the usable frequency is lower than for small-capacity capacitors. In addition, it matters which type of capacitor you use. Ceramic or mica capacitors are usable up to a much higher frequency than an MKT polyester capacitor.

At a high frequency, the peaks of the sine wave are closer together in time than at a low frequency, therefore the condenser for high frequencies may be smaller than that for low frequencies to achieve the same smoothing effect.

To make the single-phase rectifier function over a wide frequency range, two capacitors are connected in parallel a polyester capacitor with relatively high value for the low frequency range and a ceramic or mica capacitor with relatively low value for the high frequency range.

Because the DC load of the rectifier is high and draws little current, the values of the two capacitors are not critical. In practice, the following values are satisfactory.

- C1 = 100 pF or higher, ceramic, mica or similar
- C2 = 47 nF, MKT or similar

### Pi filter

To prevent HF getting onto the DC output pins, a Pi filter with a coil can be used Because I had the necessary components lying around, I built this project with PI filter, but this is not strictly necessary. The meter also works very well without a Pi filter. The value of the coil is not critical, I chose the same value that Mile Z33T uses in his design.

- C1 = 100 pF or higher, ceramic, mica or similar
- C2 = 47 nF, MKT or similar
- C3 = 100 pF or higher, ceramic, mica or similar
- C4 = 47 nF, MKT or comparable
- L = 470 μH

### Diode specifications

For this design, I used the BAT41 Schottky diode, this is a fast switching diode with low knee voltage, high maximum reverse voltage and low reverse capacitance. This diode can be bought on several major online platforms. In the Netherlands, this diode is supplied by e.g. HaJé

Type | Technology | Uk = Knee tension | Ur = Max reverse | Cr = Reverse capacity |
---|---|---|---|---|

BAT41 | Shottky | 0.3 Volt | 100 Volt | 2pF |

### Number of diodes

The diode BAT41 used has a maximum reverse voltage of 100 volts. With those given above formules , it is possible to calculate how many diodes need to be connected in series, for measuring higher powers.

The dummy load used it a maximum power of 250 Watts

`P` | `Urev` |
---|---|

250 W | 316 V |

This is just above the maximum `Urev` of 300 volts of three diodes in series.

Calculation of maximum measurable power using `Urev` for 3 and 4 diodes in series.

Number of diodes | `Urev` | `Pmax` | `Uout max` |
---|---|---|---|

1 | 100 V | 24,8 W | 49,85 V |

2 | 200 V | 99,4 W | 99,7 V |

3 | 300 V | 223 W | 149 V |

4 | 400 V | 397 W | 199 V |

5 | 500 V | 621 W | 249 V |

Since for me the upper limit of 223 W is more than enough, I opted for a design with three BAT41 diodes. If you want to be able to measure more power, choose four or more diodes.

When using multiple diodes, the overall reverse capacitance is reduced and the frequency range is significantly increased.

The total reverse capacitance equals:

`CrevT = (2 pF)/N`

`N` = number of diodes

The number of diodes affects the minimum power to be measured. This is only relevant when measuring powers in the milliwatt range.

Number of diodes | `Ukt` | `Pmin` |
---|---|---|

1 | 0,3 V | 0.9 mW |

2 | 0,6 V | 3.6 mW |

3 | 0,9 V | 8.1 mW |

4 | 1,2 V | 14.4 mW |

5 | 1,5 V | 22.5 mW |

### Output resistance

Finally, i added i my design an output resistor of 10 kΩ, this resistor limits the output current when the output is short-circuited and protects the diodes from overload.

From the BAT41, the maximum allowable If = is 100 mA, provided the connecting wires remain at room temperature. It is wise to keep the current at shorted output well below this value to avoid unnecessary heat development in the diode. With a resistor of 10 kΩ, the following maximum If can be generated.

`Uo` | `If` |
---|---|

150 V | 15 mA |

200 V | 20 mA |

This is well below the maximum If = 100 mA of the BAT41 and provides effective short-circuit protection. A practical measurement with a Digital Multimeter shows Uo in front of and behind the diode to deviate less than 1%. This is well within accuracy I want to achieve in this project.